Prediction of Smoke Propagation: Bringing Global Awareness


SmokeAlert team

Awareness of people and emergency services could diminish consequences of smokes

Our approach - technical concept

Down arrow

Fires-map app for Siberia, July 2016

- open map standalone

Smoke plumes over Siberia, Russia in july 2016

MODIS bands (RGB:3-6-7, 500 m/px), mask detection - SmokeAlert

Consumer services demo

Conclusion

During the hackathon we have:
  • Met each other!
  • Identified and verified the need for the service described
  • Discovered data sources for the model and implemented data fetcher
  • Chose the last year fires in Siberia, Russia as a pilot region for model training
  • Built the heuristic to classify smoke
  • Developed requirements for a spaciotemporal prediction model
    (+ density-based clustering for hotspots)
  • Developed a prototype of web-map with data layers and alerts
  • Developed mockup of mobile prototype to show the case of mobile alerts
  • Prepared visual demo

Our team

  • Georgy Potapov, creator of fires-map.com, Crisismap
  • Dmitry Pogodin, backend-developer
  • Maria Kolos, UX / UI developer
  • Oleg Urzhumtsev, data scientist, Skoltech, Cisco
  • Rustam Akhtyamov, PhD-student, Skoltech Space Center